The news: Retailers exploring agentic AI are approaching it with clear commercial goals, not just curiosity. In a recent EMARKETER interview, Criteo platform strategy vice president Michael Greene said retailers view the onsite shopping experience as “mission critical” and want AI that “creates superior shopping experiences within their individual category” by drawing on their own product expertise and data.

Greene also noted that retailers increasingly recognize the limits of generic LLMs: “Certainly if [a retailer’s experience] surpasses what you could get from a product recommendation from a generic LLM … all the more reason you’d keep going back there,” he said.

Why it matters:

• Retailers face pressure as consumers adopt chat-based shopping tools. Greene explained that retailers have long “been fighting … to own the shopper” and now must also compete with LLM interfaces for product discovery.
• A sizable share of shoppers already use AI for foundational research, with 38% turning to AI to learn about a brand, and 36% using it to explore a product or category, per Future Commerce. This reinforces Greene’s point that retailers must meet shoppers earlier in the journey with better answers than a general LLM can provide.
• Gen Z is the most AI-forward shopping cohort: 61% have used AI to help with a purchase in the past year, compared with 57% of millennials, per Paypal. This underscores Greene’s point that retailers must deliver AI experiences good enough to keep younger shoppers onsite rather than drifting to general-purpose chat interfaces.